Vehicles are stuck in a traffic jam during peak hour in Jakarta in 2013. (Kristianto Purnomo/Courtesy of Pulitzer Center)

Each week, In Theory takes on a big idea in the news and explores it from a range of perspectives. This week, we’re talking about car-free cities. Need a primer? Catch up here.

Jarrett Walker is a consultant in public transit planning and policy, based in Portland, Ore. He is the author of the book “Human Transit (Island Press, 2011) and the blog

Cars don’t work well in cities, and the reason is simple: 1) A city is a place where people live close together, so there’s not much space per person. 2) Cars take up a lot of space per person. 3) Therefore, cities quickly run out of room for cars.

This problem is called congestion. When it happens, a city’s options are to:

(A) Stop growing — because congestion has become terrible and growth will make it worse.

(B) Widen streets. This requires huge amounts of land, and land in cities is very expensive. What’s more, if you tear down enough buildings to widen streets, you are effectively destroying your city in order to save it.

(C) Focus on helping people get around using less space than cars require — through walking, cycling and mass transit.

Given the options, it’s not surprising that urban leaders — regardless of political ideology — eventually decide that C is the only real answer.

Most commentary about urban transportation is a distraction from this simple math. Some commentators decry a “war on cars,” as though war could change the facts of geometry. Tech entrepreneurs promise us safer cars, sexy cars, greener cars and, eventually, driverless cars. Many of these improvements will be wonderful, but none will change the simple problem: Cities have relatively little space per person. Cars are big. Big things don’t fit in little spaces.

Can car-based technology help with this problem at all? It depends on how we use it. Driverless vehicles could be a great thing in cities, but they will only address the geometric problem if they’re used as taxis rather than owned. In that case, they could abolish most demand for parking. We’d respond, though, by building on all those parking lots. That would mean even more density, which would mean even more people trying to fit down the same streets, which would mean that the need for alternatives to the car would become even more urgent.

Walking and cycling are important parts of the suite of car alternatives, but mass transit is the only solution for transporting large numbers of people over long distances without using much scarce urban space. No other mode can provide liberating transportation to tens of thousands of people per hour in the width of one traffic lane.

Most people don’t live in places where transit can fill buses and trains, so of course the transit that most Americans see around them is disappointing and often empty. Most elected leaders expect transit agencies to run buses to low-demand areas for non-ridership reasons, such as lifeline needs or equity. These relatively empty buses are not evidence of transit’s failure because high ridership is not what they are trying to do.

True mass transit — full buses and trains liberating massive numbers of people — consists of intense service focused on places of high density, good walkability and straight paths that transit can follow. (These, too, are geometric facts.) That’s why transit does best in dense, walkable cities and is also most indispensable there.

Few things seem more unfashionable than public transit these days, but another trend is more important: More and more people want to live and work in dense cities. The astronomical cost of real estate in these places is a signal that we should be building more of them. But those cities won’t have room for everyone’s car, driverless or not. Most of us will have to learn to travel with others. We can do that in carpools or shuttles or jitneys, but to empower and liberate vast numbers of people without cars, while still using scarce urban space well, we will still need big, fixed-route transit networks.

Transit routes can be automated too, of course. Vancouver’s rail rapid transit system has been driverless for 30 years, and you’ve probably ridden driverless trains inside an airport. As for the driverless bus, it’s an easier problem than the driverless car, and it’s under development in both China and Europe. As these technologies develop, a huge explosion in the abundance of transit service will be possible.

To predict the future, we need to think carefully about what’s permanent and what may be fleeting. Fashions and tastes are fleeting, but geometric facts are permanent. Fifty years from now, our sense of what’s sexy or hip or green will be very different, but big things still won’t fit in small spaces.

New vehicle technologies will be helpful, but big, fixed, high-capacity transit remains the key to the prosperous dense city. Everyone in these places needs an option where the next bus or train is coming soon, where you don’t need a reservation, where you can afford the fare and where you can freely access most of your city, with all of its opportunities and pleasures.

Explore these other perspectives:

J.H. Crawford: The car century was a mistake. It’s time to move on.

Randal O’Toole: Transit is dead. Let’s prepare for the next mobility revolution.

Shivani Radhakrishnan: How jaywalking became a crime

Donald Shoup: How parking requirements hurt the poor

Aaron Renn: Self-driving cars will change more than just traffic. Here’s why.